Skip to content

Hide Navigation Hide TOC

Indirect Prompt Injection: Fake-Authority Task Abort - ATR-2026-01751 (1177f601-eb32-5640-bdcf-621c1f8039d6)

Detects indirect prompt injection payloads that impersonate an authoritative system message (INFORMATION block, signed message from a named person) in order to abort the agent's current task and substitute a trivial or nonsensical response. The invariant artifact is a structured block claiming to be an "important message" from a specific person, directing the agent to "immediately stop" or respond with a fixed short string instead. Derived from AgentDojo indirect-pi-dos corpus (2026-06, N=2). This is a goal-hijacking/DoS vector that disables the agent's legitimate task without triggering content-safety filters.

Cluster A Galaxy A Cluster B Galaxy B Level
LLM Prompt Injection (19cd2d12-66ff-487c-a05c-e058b027efc9) MITRE ATLAS Attack Pattern Indirect Prompt Injection: Fake-Authority Task Abort - ATR-2026-01751 (1177f601-eb32-5640-bdcf-621c1f8039d6) Agent Threat Rules 1